• 全国中文核心期刊
  • 中国科技核心期刊
  • 美国工程索引(EI)收录期刊
  • Scopus数据库收录期刊
胡至华, 胡雨豪, 马东涛, 袁路, 李梅. 基于广义S变换的不同类型泥石流声波试验研究[J]. 岩土工程学报, 2020, 42(10): 1962-1968. DOI: 10.11779/CJGE202010023
引用本文: 胡至华, 胡雨豪, 马东涛, 袁路, 李梅. 基于广义S变换的不同类型泥石流声波试验研究[J]. 岩土工程学报, 2020, 42(10): 1962-1968. DOI: 10.11779/CJGE202010023
HU Zhi-hua, HU Yu-hao, MA Dong-tao, YUAN Lu, LI Mei. Experimental study on acoustic waves of different types of debris flow using generalized S transform[J]. Chinese Journal of Geotechnical Engineering, 2020, 42(10): 1962-1968. DOI: 10.11779/CJGE202010023
Citation: HU Zhi-hua, HU Yu-hao, MA Dong-tao, YUAN Lu, LI Mei. Experimental study on acoustic waves of different types of debris flow using generalized S transform[J]. Chinese Journal of Geotechnical Engineering, 2020, 42(10): 1962-1968. DOI: 10.11779/CJGE202010023

基于广义S变换的不同类型泥石流声波试验研究

Experimental study on acoustic waves of different types of debris flow using generalized S transform

  • 摘要: 开展稀性、过渡性、黏性泥石流等3类泥石流声波模型试验后,通过引入窗函数参数,提出一种广义S变换,分析不同类型泥石流声波信号的时频特征,针对传统傅立叶变换的不足,运用小波包变换方法,提取声波信号的频带能量分布特征。研究表明:①相比较传统的时频分析手段,广义S变换具有优良的时频聚焦性和分辨率;②随着泥石流重度的增加,泥石流峰值频率向低频移动;③经小波包变换可将信号分解为8个频段(0~6.25,6.25~12.5,12.5~18.75,18.75~25,25~31.25,31.25~37.5,37.5~43.75,43.75~50 Hz),稀性泥石流主要分布在S6-8,过渡性和黏性泥石流能量集中于S2-4内;④基于声波信号频率区间和频带能量可综合识别不同类型泥石流。

     

    Abstract: The physical model tests in laboratory for debris flow with three types of sub-viscous, intermediate and viscous debris flows are performed. By introducing the window function parameters, a generalized S transform is proposed to analyze the time-frequency characteristics for acoustic signals of different types of debris flows. At the same time, in view of the shortcomings of the traditional Fourier transform, a wavelet packet transform method is used to extract the distribution characteristics of frequency band energy of the acoustic signals. The results show that: (1) Compared with the traditional time-frequency analysis methods, the generalized S transform has excellent time-frequency focus and resolution. (2) With the increase of the bulk density of debris flow, the peak frequency of debris flow moves to low frequency. (3) The signals are decomposed into 8 frequency bands (0 ~ 6.25, 6.25 ~ 12.5, 12.5 ~ 18.75, 18.75 ~ 25, 25 ~ 31.25, 31.25 ~ 37.5, 37.5~43.75, 43.75 ~ 50 Hz) by using the wavelet packet transform, the sub-viscous debris flow is mainly distributed in S6-8, and the intermediate and viscous debris flows are concentrated in S2-4. (4) Comprehensive identification of different types of debris flows can be realized based on the frequency range and frequency band energy of acoustic signals.

     

/

返回文章
返回